University of Virginia Library

III

As Greg remarked, the second section of the book (sheets H through L) is set in a slightly different letter from that used in the first section, a fact which suggests that it was composed by a workman different from the two working there. Unfortunately in this section there are few of the typographical peculiarities of the kind found in the first; nevertheless, there is enough evidence available to permit an estimate of the probable order of the setting of the formes.

The pattern of running titles indicates that, with the exception of sheet L, the method of composition employed in the second section was the same as that employed in the first. Unless a press delay is hypothecated, the order of imposition (and probably of composition also) must have been either:

or In the last line of H3 there is a distinctive capital T (in the word "The") which reappears in the word "To" on I3 (l.19). Had the inner forme of H gone to press before the outer, it is unlikely that the outer could have been set, imposed, wrought off, and distributed before the composition of I3; therefore, it seems reasonable to reject the second order shown above and to accept the first. The inverse order of the composition of the formes of sheet L can be accounted for if it is supposed that the compositor wished to take advantage of an opportunity to get a forme ready for the press somewhat earlier than usual: since L4v was blank, L(o) could be imposed after the setting of only three type pages rather than four.

The analysis of the printing of the first section of the book indicated that the method of composition by the prior setting of alternate formes of succeeding sheets could be employed by one compositor. To ascertain whether or not there was indeed only one man at work in sheets H through L and whether S's or T's characteristics appear there, spelling evidence must again be adduced. A total of thirty-one variant spellings in the play was tested by the method advocated by Hinman[22] and used by such investigators as Williams[23] and Brown.[24] Of the thirty-one words, eight were found to be significant, but of this number some were of greater value than others for the purpose at hand. The pattern of their occurrence indicates that the

last four sheets of the book were probably set by a single workman (designated Compositor U) who had different spelling characteristics from those of Compositors S and T. The evidence may be summarized as follows:                                  

	B-F/G3	G1-2v/G3v-4v	H-L
againe	13	1	10
agen	5	5	9
blood	13	4	0
bloud	0	0	9
deere	5	0	1
deare	3	1	4
eye	17	4	3
eie	6	0	5
safty/safety	2	1	0
safetie	1	0	4
sweete	5	3	1
sweet	2	0	3
to (i.e., also)	0	2	0
too	0	1	8
yeare	7	2	1
yeere	0	0	2

From this tabulation it is clear that the case for the presence of Compositor U rests largely on the spelling of bloud as opposed to blood. The other spellings, although weak as evidence in themselves, tend to confirm the division suggested by the bloud spellings; the presence in sheet H of deare, eie, and safetie permits the addition of this sheet to U's stint even though bloud does not occur there. A cluster of againe and eye spellings in sheet L is somewhat disconcerting since both are S's preferred forms, but L also contains three bloud spellings and for this reason the entire sheet is assigned to U. Indeed, the consistency of the typography of the last four sheets and the spelling evidence lead to the conclusion that the second section was set without assistance by one compositor. He is distinguished from S and T by his strong bloud preference and by less pronounced preference for eie, safetie, sweet, and yeere. He differs from T alone in showing no preference for either agen or againe, in preferring honour to honor (his stint contains seventeen honour and two honor spellings), in preferring too to to, and possibly in preferring young to yong (there is one young on L1, while yong does not appear).

Greg's suggestion that The Maid's Tragedy Q1 may have been printed in two shops has been noted above. Compositor S was clearly Okes's man

because on B1 he used an ornament and an initial known to have belonged to Okes. The work of Compositor T is linked to that of S by his use of skeleton formes which were previously used by S and his sharing of the same font of small capitals. But U used different running-titles and set from a different case. Nevertheless, I believe that there is ample evidence to show that he worked in Okes's rather than another shop. First, the roman letter of sheets H through L of The Maid's Tragedy is the same as that used in Daniel's Whole Workes (STC 6238), which was printed by Okes in 1623. Second, the type of the act head which appears on sig. H4v of The Maid's Tragedy appears to be of the same font as the act heads used in Philaster Q2, which was printed by Okes in 1622. Third, the running-titles of both sections of The Maid's Tragedy, although different settings, appear to be from the same font, which was also used by Okes for the composition of the headlines of Thierry and Theodoret Q1 in 1621. Last, U's preferred spellings, bloud and eie, appear in The Honest Whore Q4, a one-compositor play printed by Okes in 1616. One may conclude that The Maid's Tragedy Q1 was printed entirely in Okes's shop, and it also seems probable that a delay occurred between the printing of the first and second sections of the book.

Before leaving this examination of the printing of Q1, there is one further matter of importance which requires consideration. The method used for printing a book was, as it has been pointed out, closely related to the method used for obtaining proof. Since The Maid's Tragedy Q1 was composed and printed in an unusual way, it seems possible that the proofing might also have had some unusual features.

The essential difference between one- and two-skeleton printing, as far as proofing was concerned, lay in the increased efficiency of operation which resulted from the proofing of forme II of sheet X while corrections were being made to the type of forme I.[25] But if a book was set by formes, and, as in the case of The Maid's Tragedy Q1, forme I was imposed and sent to the press before the composition of forme II began, the machining of forme I would have been completed at about the same time as the locking-up of forme II, provided no great disparity between the speed of composition and the speed of presswork existed. Under these circumstances forme II could not have been ready to go on the press when forme I was removed for correction. The only advantage gained from using two skeletons was that forme II could be imposed when forme I was still on the press; the taking of proof and the correcting of the type must have been done exactly as if only one skeleton were being employed.

There is every reason to believe that press delays were abhorrent to the 17th-century printer. One-skeleton proofing procedure necessarily caused the press to wait while corrections to the type were being made. Therefore, although there is no evidence bearing directly on the point, it seems likely

that the proofing of a one-skeleton book, or a two-skeleton book printed in such a way as to cause one-skeleton proofing procedure to be adopted, may have been a more hurried, and perhaps a more casual, affair than the proofing of a two-skeleton book.[26] At any rate, the proofing of The Maid's Tragedy Q1 seems to have been haphazard.

Collation of the six extant copies of The Maid's Tragedy Q1 reveals the following press variants: Extant copies: Bodl. (Bodleian Library), CSmH (Henry E. Huntington Library), DFo (Folger Shakespeare Library), Dyce (Dyce Collection of the Victoria and Albert Museum), MB (Boston Public Library), and MH (Harvard University).

Sheet B (outer forme)

• Uncorrected: Bodl. CSmH
• First stage corrected: DFo
• B3 wedding ] weding Strato ] Steat
• Second stage corrected: Dyce, MB, MH
• B4v liue ˄ away ] liue, away comming in ] come in

Sheet C (outer forme)

• Uncorrected: DFo
• Corrected: Bodl., CSmH, Dyce, MB, MH
• C2v solemne ] s lemne Darke night˄] Darke night, Second ] second
• C3 aboue ] aboae contemnes ] contems
• C4v teach you ] teachyou

Sheet E (outer forme)

• Uncorrected: DFo, MB, MH
• Corrected: Bodl., CSmH, Dyce
• E1 Nor I ] Nere I

Sheet L (inner forme)

• Uncorrected: CSmH
• Corrected: Bodl., DFo, Dyce, MB, MH
• L1v lookes ] bookes
• L2 a way ] away

These corrections may be described:

It seems evident that, in spite of the delay, Okes was not unwilling to stop his press to make minute corrections: indeed, one or two of them may even have been overzealous. On the other hand, nearly every forme in the book, including the formes with variants, contains other and more obvious errors, such as the following (the lemmata are those of a hypothetical corrected state):[27]

                                                       

B(o)	B1	1	sir, ] sir.
		20	forbad ] fotbad
B(i)	B3v	27	there, codes, codes.] there, [space] codes, codes
	B4	6	this ] rhis
C(o)	C1	1	King, Evadne ] King ˄ Evadne
		15	day. ] day,
	C4v	4	credulous ] credulons
C(i)	C2	6	vernall ] veranll
		25	floud ] flould
	C4	18	a side ] aside
D(o)	D1	35	luster ] lnster
	D3	9	Instruct ] Instant
		13	thunder ] thundet
D(i)	D1v	22	againe, ] againe.
E(o)	E1	2	Olimpias ] Olimpas
		18	more ˄ pittying ] more, pittying
	E3	26	pritiest ] prtitiest
	E4v	10	royaltie ] rioyaltie
E(i)	E2	7	a bout ] about
	E4	10	honest ] honost
F(o)	F2v	21	to kill me ] can kill me
		33	a little ] alittle
F(i)	F2	16	another ] an other
G(o)	G4v	7	Safer ] Safer
H(i)	H2	35	a faith ] A faith
	H3v	28	Ile be sworne ] I besworne
I(o)	I3	26	oth' King ] oth' the King
L(o)	L2v	13	farwell, and ] farwell, And

If Okes was willing to make stop-press corrections of the kind revealed by the press variants, it seems odd that errors of the same order, as shown in the above list,[28] were allowed to stand. Had the proofreading been carefully done according to the standards suggested by the press variants, it seems unlikely that many of these mistakes would have gone uncorrected.

One may conclude, then, that the reading of the first impression to be pulled from the uncorrected formes, the "proof," was done hastily and

incompletely.[29] A few errors, some almost trivial, were noticed and corrected, probably without removing the forme from the bed of the press.[30] The total effect of the proofreading on the text of Q1 seems to have been very slight; the compositors were almost entirely responsible for the text as it now stands.